PowerAI-Diabetes: Review of glycemic and lipid variability to predict cardiovascular events in Chinese diabetic population

Abstract

The aim of this study is to review the predictive value of visit-to-visit variability in glycaemic or lipid tests for forecasting major adverse cardiovascular events (MACE) in diabetes mellitus. Data from existing studies suggests that such variability is an independent predictor of adverse outcomes in this patient cohort. This understanding is then applied to the development of PowerAI-Diabetes, a Chinese-specific artificial intelligence-enhanced predictive model for predicting the risks of major adverse cardiovascular events and diabetic complications. The model integrates an amalgam of variables including demographics, laboratory and medication information to assess the risk of MACE. Future efforts should focus on the incorporation of treatment effects and non-traditional cardiovascular risk factors, such as social determinants of health variables, to improve the performance of predictive models.

Fig. 1. The clinical consequences and biological mechanisms of glycemic variability in relation to MACE events.

Reproduced from ref. with permission.

Fig. 2

Summary of pathogenic mechanisms underlying glycemic/ lipid variability and major adverse cardiovascular events.

Fig. 3. Approach for model development.

Our model can accurately predict nine different diabetes-related complications (neurological, ophthalmological, CKD, dementia, osteoporosis, peripheral vascular disease, ischemic heart disease, atrial fibrillation and heart failure). Reproduced from ref. with permission.

Fig. 4

Dashboard showing estimation of 3-, 5- and 10-year risk of heart failure in a high-risk patient (top) and a low-risk patient (bottom).

Fig. 5

Integration of massive amounts of data from different domains to develop state-of-the-art models enhanced by artificial intelligence.